Browse this article:
We evaluate each camera by taking a selection of photos in controlled conditions with the help of a tripod, and we also take some hand-held general-purpose shots - the kind most people take in real life. We use the cameras in a wide range of situations, from bright outdoor conditions to very dark indoor scenes. We also use all the functions available on the cameras to get a good idea of their flexibility and usability.
As the seven cameras on test are mainly aimed at the amateur point and shoot market, we use them in automatic mode to assess how well they meter and expose images in varying conditions. This also tests the effectiveness of the automatic white balance and noise-reduction systems. Most of the cameras include continuous shooting and movie modes, which we also try out. However, as these cameras are mainly used to shoot single stills, we concentrate on this area in our tests.
Most of the photos taken on these cameras will be of people, so we take two indoor shots under the fluorescent lighting in our office. We used a tripod to ensure compatibility across images from different cameras, but as each camera differs we had to adjust the tripod's angle and degree of rotation. The exact same distance was maintained.
We try to frame the shot identically for each camera and use both forced and suppressed flash. Where the auto white balance fails to produce acceptable colours, we note this in the individual reviews. Photos were taken at widest maximum, full optical zoom and at full digital zoom. Each shot was performed with forced and suppressed flash.
We take the majority of our outdoor test shots in real world conditions, mainly basing our final rating on the resulting collection. For an apples to apples comparison, we took a shot of a silver car in bright, direct sunlight. This offers a good test of each camera's ability to resolve detail, as well as allowing us to detect any barrel distortion and colour fringing. We also look at how well each camera exposes the image.
What we look for
|Colour fringing (misregistration) is clearly visible in the purple line running along the edge where white meets black.|
After uploading the images onto our usual test rig, we scrutinise them for issues like barrel distortion, colour fringing and tonal range. We also look for general colour accuracy, resolution and errors such as noise and compression artefacts. Shots with flash are checked for the spread and range of the flash and again for colour balance errors. Outdoor photos are examined for accurate tones, exposure, colour and detail. We calculate the area covered in the macro shot and check focus across the image, particularly in the corners.
How we calculate the final scores
As with every PC Authority group test, we carefully allocate points to every feature, with the more important aspects getting larger scores. For the cameras here, this covers everything from optical zoom ranges and bundled accessories to shooting modes, controls and memory. Lastly, we take usability into account. This covers the cameras' physical controls, menu layout and shutter lag.
|Barrel distortion causes a bending of straight subjects on the edge of captured images. The bar (left) should be straight.|
After adding up all the points, we convert them into a weighted average for all of the cameras, where 100 is bang on average.
Performance is derived from the quality of all the shots taken. We take a weighted average of all images, with scores for image aspects such as clarity, colour, contrast, flash and more. This is again converted into a score out of six.
Value for money is calculated by multiplying the price by a combination of 50 percent features and design and 50 per cent quality. This figure is then weighted again to give a weighted average value for money rating.
Finally, the overall score is calculated as a straight average of performance, features and design, and value for money.
Browse this article:
This Group Test appeared in the January, 2005 issue of PC & Tech Authority Magazine